Seamless Cluster Service Migration via Subgroup Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer systems face challenges in efficiently migrating and upgrading services across nodes, particularly in distributed environments, where stateful and stateless clusters require seamless transitions with minimal disruption and resource overhead.
Innovation Solution
The approach involves progressively migrating source computer nodes by grouping them into subgroups, scheduling the creation of target subgroups, and activating them only after deactivating the source subgroups, allowing for efficient resource reuse and minimal disruption during software upgrades or service migrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If services are migrated from source computer nodes to target computer nodes in a distributed environment, then service continuity and minimal user disruption are achieved, but system complexity and resource overhead increase
Solution Approach 1:
The patent segments the computer nodes into source subgroups and target subgroups, allowing incremental migration of services from source to target nodes. This segmentation enables controlled migration where services are transferred in manageable units rather than all at once, maintaining service continuity while managing system complexity through phased transitions.
Solution Approach 2:
The patent creates target subgroups before deactivating source subgroups, ensuring that target nodes are ready to immediately assume services when source nodes are deactivated. This preliminary preparation of target infrastructure prevents service disruption and maintains reliability during the migration process.
2Productivity
If source computer nodes are deactivated and target subgroups are created, then resource reuse and efficiency are improved, but migration time and operational complexity increase
Solution Approach 1:
The patent implements periodic migration by deactivating source subgroups and activating corresponding target subgroups in sequential phases. This periodic approach allows the system to migrate services in structured intervals, enabling resource reuse while managing migration time through controlled, incremental transitions rather than attempting complete migration simultaneously.
Solution Approach 2:
The patent deactivates source subgroups after their services have been successfully migrated to target subgroups, thereby recovering and releasing the source nodes for reuse. This discarding and recovering mechanism improves resource efficiency by making previously occupied nodes available for new assignments while maintaining service continuity through the target subgroups.
3Quantity of substance
If target subgroups are created only after source subgroups are deactivated, then resource overhead is reduced, but service migration flexibility and responsiveness decrease
Solution Approach 1:
The patent implements dynamic migration by allowing the system to adapt the timing and sequencing of target subgroup creation based on service requirements and system state. The scheduler can dynamically adjust when to create target subgroups relative to source subgroup deactivation, providing flexibility in managing resource overhead while responding to changing service demands and maintaining migration adaptability.
Data Source
AI summary
Embodiments are directed to progressively migrating source computer nodes where the source computer nodes perform a computer-implemented service. In one embodiment, a computer system determines that execution of the performed service is to be migrated from the source computer nodes to target computer nodes. The computer system groups the source computer nodes into multiple source subgroups, where each source subgroup includes at least one source computer node. The computer system then schedules creation of target subgroups of target nodes. These target subgroups include at least one source computer node and, themselves, correspond to a source subgroup. The computer system activates a first target subgroup corresponding to a first source subgroup, and deactivates the first source subgroup. In this manner, the first target subgroup replaces the first source subgroup. Still further, the target subgroups are scheduled to be created only after the first source subgroup has been deactivated.


